© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 32-TRANSCODER-HP
    7. 43746
    Prev
    Next
    INDEX
    Explanations

    military/police operations

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    eton
    -0.28
    pawn
    -0.27
    sville
    -0.25
    å¯Įæ°ij
    -0.25
    æĪ¾
    -0.25
    /activity
    -0.25
    ss
    -0.24
     endors
    -0.24
    activate
    -0.24
    haul
    -0.24
    POSITIVE LOGITS
     searches
    0.33
     search
    0.29
    VIP
    0.28
     mine
    0.27
    ENU
    0.27
     VIP
    0.26
     support
    0.25
     direct
    0.25
     court
    0.25
     law
    0.25
    Activations Density 0.004%

    No Known Activations